Project Summary: Development and testing of disease-modifying therapies for Parkinson's disease (PD) is limited, in part, by lack of predictive animal models. We propose to develop and characterize a better, more predictive model, with (i) a more realistic time course, (ii) appropriate pathological and behavioral endpoints and (iii) opportunities to intervene therapeutically at clinically relevant points in the disease course (e.g., after development of symptoms). We believe the new model will set a higher and more realistic hurdle for experimental, disease-modifying therapies, and in so doing, will have more predictive value for clinical development. In brief, wildtype rats are treated with rotenone for 5 days (instead of the typical 14-21 days), during which they displayed mild parkinsonian behavior. After cessation of the rotenone, the rats recover and are behaviorally normal. After a latency of 9 ? 10 weeks, all of the rats begin to exhibit mild parkinsonian signs that have continued to worsen progressively over succeeding weeks and months. Preliminary pathological assessment reveals that nigral dopamine neurons progressively accumulate abnormal endogenous ?-synuclein long before behavioral signs appear and this is associated with increasing microglial activation. We now propose to further characterize this model and to test a therapeutic intervention with the following aims: (1) To characterize selected `clinical' or behavioral aspects of this model over the course of 1 year. Behaviors include the Postural Instability Test (PIT), rearing, righting behavior and gait. (2) To characterize over time the pathology in this model, including nigrostriatal cell and terminal loss, ?-synuclein pathology and markers of inflammation. (3) To test a therapeutic intervention (?-synuclein knockdown) after the onset of symptoms ? a situation closely analogous to clinical medicine, where a diagnosis of PD currently depends on the presence of symptoms.